A short blog post that strings together 8 tweets that I sent out today about our new paper.
Today our latest paper on grammatical aspect appeared in Collabra: Psychology. The article reflects the times we psychologists are living in. It does so not from the lofty perspective of the methodologist or statistician, but from the work floor on which the actual scientist (**ducks**) operates.
Our first two experiments were inspired by Hart & Albarricin (2011). This research itself was inspired by some of our own work but took it from cognition into the realm of social psychology, as I described in this blog post.
As the paper explains, these experiments were run in 2012, which is why they were not preregistered. Nobody was doing preregistration at the time. We were thinking to build on Hart and Albarricin (H&A) in what some would call a conceptual replication but which is better thought of as an extension.
For the life of us, we couldn’t get an effect like that of H&A. Then we got down to business and started a registered replication project in which we performed a direct replication of H&A. Along with 11 other labs, we found no effect.
We were sidetracked by the replication project. Especially because there were some troubling issues with the initial response to our RRR, as I describe here . We were sidetracked to the point that I’d completely forgotten about our 2012 experiments.
Luckily my co-authors had not and we decided to pick up the pieces of our study. It was clear that our research could no longer be driven by our H&A-inspired hypothesis, so we took a slightly different tack.
We conducted three more experiments, now all pre-registered, which yielded some interesting new findings, which you can read about in our paper. As usual per Collabra, the data are available and the reviews are open.
Today our latest paper on grammatical aspect appeared in Collabra: Psychology. The article reflects the times we psychologists are living in. It does so not from the lofty perspective of the methodologist or statistician, but from the work floor on which the actual scientist (**ducks**) operates.
Our first two experiments were inspired by Hart & Albarricin (2011). This research itself was inspired by some of our own work but took it from cognition into the realm of social psychology, as I described in this blog post.
As the paper explains, these experiments were run in 2012, which is why they were not preregistered. Nobody was doing preregistration at the time. We were thinking to build on Hart and Albarricin (H&A) in what some would call a conceptual replication but which is better thought of as an extension.
For the life of us, we couldn’t get an effect like that of H&A. Then we got down to business and started a registered replication project in which we performed a direct replication of H&A. Along with 11 other labs, we found no effect.
We were sidetracked by the replication project. Especially because there were some troubling issues with the initial response to our RRR, as I describe here . We were sidetracked to the point that I’d completely forgotten about our 2012 experiments.
Luckily my co-authors had not and we decided to pick up the pieces of our study. It was clear that our research could no longer be driven by our H&A-inspired hypothesis, so we took a slightly different tack.
We conducted three more experiments, now all pre-registered, which yielded some interesting new findings, which you can read about in our paper. As usual per Collabra, the data are available and the reviews are open.
Reacties
Een reactie posten